> On June 12, 2015, 11:20 p.m., Jie Yu wrote: > > Have you tested this?
Yep, the test was updated to properly test this, and I verified manually. :) > On June 12, 2015, 11:20 p.m., Jie Yu wrote: > > 3rdparty/libprocess/src/process.cpp, line 593 > > <https://reviews.apache.org/r/35404/diff/1/?file=983948#file983948line593> > > > > Not sure if this will cause a scalability issue since we are calling > > getpeername everytime there are some requests coming in. Maybe we should > > cache the peer address when the socket is accepted. Can you add a TODO? Added a TODO on Socket::peer which matches the existing one in Socket::address. FWIW I measure about 1 microsecond for this call, excluding timer overhead, so should be ok without the caching, but I'd love to avoid the syscall overhead here! :) - Ben ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/35404/#review87781 ----------------------------------------------------------- On June 12, 2015, 10:40 p.m., Ben Mahler wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/35404/ > ----------------------------------------------------------- > > (Updated June 12, 2015, 10:40 p.m.) > > > Review request for mesos and Jie Yu. > > > Bugs: MESOS-2519 > https://issues.apache.org/jira/browse/MESOS-2519 > > > Repository: mesos > > > Description > ------- > > This now uses `Socket::peer` to augment the `http::Request`. > > > Diffs > ----- > > 3rdparty/libprocess/src/process.cpp > aadd7bb0ae12b93336900c76d8d5aaa4421ea198 > 3rdparty/libprocess/src/tests/http_tests.cpp > 8444b9c961d04b932188b2ac37e2a42aafda1abd > > Diff: https://reviews.apache.org/r/35404/diff/ > > > Testing > ------- > > Updated the test. > > > Thanks, > > Ben Mahler > >
